University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln USGS Staff -- Published Research US Geological Survey 5-2000 Parent Brine of the Castile Evaporites (Upper Permian), Texas and New Mexico Walter E. Dean U.S. Geological Survey, Denver, CO, [email protected] Douglas W. Kirkland Department of Geosciences, The University of Texas at Dallas, Richardson, Texas 75288, U.S.A. Rodger E. Denison Department of Geosciences, The University of Texas at Dallas, Richardson, Texas 75288, U.S.A. Follow this and additional works at: https://digitalcommons.unl.edu/usgsstaffpub Part of the Earth Sciences Commons Dean, Walter E.; Kirkland, Douglas W.; and Denison, Rodger E., "Parent Brine of the Castile Evaporites (Upper Permian), Texas and New Mexico" (2000). USGS Staff -- Published Research. 307. https://digitalcommons.unl.edu/usgsstaffpub/307 This Article is brought to you for free and open access by the US Geological Survey at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in USGS Staff -- Published Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. PARENT BRINE OF THE CASTILE EVAPORITES (UPPER PERMIAN), TEXAS AND NEW MEXICO DOUGLAS W. KIRKLAND,1 RODGER E. DENISON,1 AND WALTER E. DEAN2 1 Department of Geosciences, The University of Texas at Dallas, Richardson, Texas 75288, U.S.A. e-mail: [email protected] 2 U.S. Geological Survey, Denver, Colorado 80225, U.S.A. ABSTRACT: The Upper Permian (lower Ochoan) Castile Formation is Prelude to the Castile Evaporites a major evaporite sequence (;10,000 km3) of calcite, anhydrite, and In late Guadalupian time (Late Permian) during an interval of tectonic halite in west Texas and southeastern New Mexico. Traditionally the quiescence in west Texas and southeastern New Mexico, a deep topograph- Castile brine has been considered to have been derived from seawater. ic basin became nearly encircled by the Capitan reef. The reef extended This tradition has recently been challenged by two versions of the along the margin of the basin for more than 500 km (e.g., Adams and closed-basin drawdown model. They call for deposition from a mixed Frenzel 1950) (Fig. 3A). Reef debris formed slopes up to 358 (King 1947; brine, in part marine and in large part nonmarine. They propose draw- Scholle et al. 1992), and the reef front had slopes approaching the vertical down of as much as 500 m to form a major sink for ground water (B.L. Kirkland et al. 1993) (Fig. 3B). Seawater is traditionally thought to issuing from the surrounding Capitan reef complex. A large fraction have entered the basin through a southern gap in the reef, the Hovey chan- of the solute in the brine body is inferred to have been recycled from nel (Fig. 3A); however, the evidence is ambiguous, and a southwestern older Permian evaporites on the surrounding shelf. channel is possible (Hill 1999). The average depth of seawater in the sed- Strontium-isotope analyses show no evidence that meteoric ground iment-starved basin was about 500 m (e.g., Newell et al. 1953, p. 190). water was contributed to the Castile brine. From a stratigraphic, geo- Landward from the reef was a wide shelf-facies complex of tidal, lagoonal, graphic, and lithologic array of 65 samples of anhydrite, gypsum, and and sabkha environments. Here the depth of seawater was seldom more calcite, 59 have an 87Sr/86Sr ratio of 0.706923 (Dsw of 2225.0), a ratio than a few meters (Fig. 3B). Carbonates accumulated directly shelfward that is the same as that of strontium in early Ochoan ocean water. If from the reef, evaporites and redbeds accumulated farther shelfward, and considerable (.15%) in¯ux of meteoric water had occurred, enough redbeds accumulated in the hinterland (Fig. 3A). continental strontium would have been introduced to have resulted in In latest Guadalupian time, shortly before Castile deposition, the channel higher ratios. (or channels) through which seawater entered the basin became restricted. Low bromide values (20±40 ppm) in Castile halite, which have been This restriction may have occurred when the Capitan reef and reef debris used to argue for meteoric in¯ux and for recycled salt, probably re- protruded into the channel (Ross 1986; Garber et al. 1989) or when the sulted from diagenesis. During shallow burial by halite, centimeter-size, channel became narrower and shallower because of a fall in sea level (For- bottom-grown crystals of gypsum were altered to nodular anhydrite. ney 1975; Ross and Ross 1987). Because of the restriction, because the The rising water of dehydration caused the halite to recrystallize. Dur- climate was arid (Kutzbach and Gallimore 1989; Scholle et al. 1992; Par- ing the recrystallization, some bromide was expelled. rish 1995), and because riverine water was absent, there was a progressive Despite the large volume of water that evaporated annually from its increase in the salinity of the basinal and shelfal marine water. surface (;52 km3/yr, assuming an evaporation rate of 2 m/yr), the Castile brine body never completely desiccated. The surrounding shelf Castile Evaporites was ¯at, hot, and generally dry. It probably could not have supplied a By the beginning of Ochoan time the reef and shelf were probably emer- signi®cant volume of meteoric spring water to the basin over tens of gent (Oriel et al. 1967; Melim and Scholle 1989; S.U. NoeÂ, personal com- thousands of years. More likely, during the entire history of the evap- munication 1998). In the basin the salinity of surface water increased until orite sequence, in¯ux was dominantly marine. Marine ground water it was several times greater than that of Permian seawater. Evaporitic pre- ¯owed through the Capitan Formation into the evaporite basin along cipitation began, and a thin (tens of centimeters) bed of laminated, silt- its southern and possibly western margin probably with a rate of ¯ow poor, aragonitic mudÐthe initial Castile evaporiteÐaccumulated on the that was usually fast enough to prevent major drawdown of the brine basin ¯oor. This bed was the harbinger of the thick sequence of calcium surface. carbonate, calcium sulfate, and sodium chloride that accumulated during the rest of Castile time (Anderson et al. 1972). The Castile Formation originally underlay at least 25,900 km2, and had INTRODUCTION a maximum thickness of about 640 m and an original volume exceeding 10,000 km3 (Adams 1944; King 1947; Snider 1966). The formation orig- Castile evaporites are exposed or shallowly buried over a large area of inally consisted, by volume, of about 5% calcite, 45% anhydrite, and 50% west Texas and southeastern New Mexico (Fig. 1). The evaporites (Upper halite (Anderson 1981), but Tertiary ground water, chie¯y along the west- Permian) represent the early part of the Ochoan stage (Adams et al. 1939). ern side of the basin, dissolved about half the halite. Siliciclastics are almost The relationship of the Castile Formation to the immediately older and absent. The Castile Formation is subdivided into a Basal Limestone Mem- younger formations with which it is associated is shown in Figure 2. The ber, four anhydrite members, I±IV (from oldest to youngest), and three most distinctive features of the Castile are small-scale, rhythmic alterations intervening halite members, I±III (from oldest to youngest) (Anderson et of calcite±gypsum, and, in the subsurface, calcite±anhydrite and calcite± al. 1972). anhydrite±halite. These alternations record a remarkably detailed meteo- The evaporite basin was probably situated between 5 and 108 N latitude rological and sedimentological record of Late Permian history (Udden (Scotese 1994; Golonka et al. 1994) on the western margin of Pangea 1924; Anderson 1982). Representative outcrops, continuous cores, and hun- adjacent to the huge Panthalassa Ocean. A monsoonal circulation, possibly dreds of boreholes constrain Castile thickness, lithology, and facies distri- analogous to that of equatorial east Africa today, led to aridity, high tem- bution, but profound differences in interpretation of its origin remain. We peratures, and high evaporation rates (Parrish et al. 1986; Kutzbach and consider here the most fundamental of the ongoing controversiesÐthe or- Gallimore 1989). The seasonal range in temperature was probably less than igin of the parent brine. 58C (Crowley et al. 1989, their ®g. 2). JOURNAL OF SEDIMENTARY RESEARCH,VOL. 70, NO.3,MAY, 2000, P. 749±761 Copyright q 2000, SEPM (Society for Sedimentary Geology) 1073-130X/00/070-749/$03.00 750 D.W. KIRKLAND ET AL. FIG. 1.ÐLocation of Castile depositional basin. Shaded rectangle is the location of map shown in Figure 6. The Castile is conspicuously cyclic, with each cycle recording progres- sively increasing salinity of the surface brine. The two most prominent cyclic units are termed ``varves''Ðlaminae (or laminae and thin beds) with a period of one year (Udden 1924), and ``millennial cycles''Ðbeds with a period of several thousand years (Dean and Anderson 1978). The most common Castile varve type consists of a coupletÐa lamina of calcite (containing organic matter) and a superjacent lamina of anhydrite (Figs. 4A, 5A). Such varves typically have a thickness of about 1.8 mm, with the lamina of calcite forming about 6% of this value. The varves are generally remarkably regular in thickness, and are commonly repeated FIG. 3.ÐA) Late Guadalupian paleogeography shortly before beginning of Castile thousands of times without interruption. The evaporitic sediments that evaporite deposition (after Ward et al. 1986). Seawater entered the basin through a formed the anhydrite±calcite couplets were cumulates. They formed at the silled channel or channels traditionally thought to have been the Hovey Channel, as brine surface or in the upper part of the brine and ``rained'' onto the basin shown. The Capitan reef front is dashed where inferred. From the basin margin, a tongue of limestone, the Lamar, extended for tens of kilometers into the deep basin.